Nova is at pains to try and prove the MWP was warmer than today (perhaps she still doesn’t realise it’s an argument for high sensitivity) and in doing so invokes a number of “science” pieces. Previously we’ve found Co2Science’s “method” was flawed and, that Huang’s boreholes didn’t contain 20th century data.
This time we examine a third piece of Nova’s “evidence”, Loehle.
In 2007 Loehle published a paper claiming …
… the MWP being approximately 0.3°C warmer than 20th century values at these eighteen sites
In 2008 Loehle published a correction and no longer made that claim, instead opting for a more cryptic one …
The warmest tridecade of the MWP was warmer than the most recent tridecade, but not significantly so.
Choose Your Own Science
In December 2009, well after Loehle had corrected 2007’s work and published the 2008 paper, Nova wrote a post titled “Fraudulent hockey sticks and hidden data”. Guess which Loehle paper she uses? That’s right, the incorrect 2007 one – ironic title to her post don’t you think! Even after she was notified by “Owen” of the later research, she declined to include it.
Only after “Nice One” recently made a comment in her blog did Nova update the post to include the new graph, however still keeps the old one under the pretence that “The updated graph makes no material difference”. Let’s compare.
Nova’s original graph looks like this …
After the paper was corrected, with data only going as far as 1935, the Loehle 2008 graph now looks like this …
The above graph only has data to 1935 and Loehle is a little ambiguous about his comparison against “20th Century Temps”.
In the 2008 paper he again produces a cryptic summary …
The warmest tridecade of the MWP was warmer than the most recent tridecade, but not significantly so. The main significance of the results here is not the details of every wiggle, which are probably not reliable, but the overall picture of the 2000 year pattern showing the MWP and LIA timing and curve shapes.
He’s saying if you average the 29 years centered around 1935, and average the 29 years around 1992, and add that difference to the end of the proxy record, then the MWP would be warmer than
today – er actually no, it would be warmer than a 29 year average from 1978 to 2006.
If you perform the same exercise but average just 11 years instead of 29, then the difference jumps from 0.4°C to 0.6°C. Even more if you just use NH GISS data, or just the last 3 years average. (Note: GISS data is used as comparison because this is what Loehle used in the report).
Proxy is Approximate
Out of interest I plotted the data from Loehle’s paper to demonstrate how discombobulated the data really is. This is no criticism of Loehle, or the data used. This is merely to show how much the data fluctuates at any given location. To get a global temp you need to at least average out the data, or better still weight it according to location and reliability.
Here’s where the samples are located. Only 3 of the 18 samples are from the Southern hemisphere, two from the bottom of South Africa and the other almost on the equator.